#value-decomposition News & Analysis

QSIM: Mitigating Overestimation in Multi-Agent Reinforcement Learning via Action Similarity Weighted Q-Learning

Researchers propose QSIM, a new framework that addresses systematic Q-value overestimation in multi-agent reinforcement learning by using action similarity weighted Q-learning instead of traditional greedy approaches. The method demonstrates improved performance and stability across various value decomposition algorithms through similarity-weighted target calculations.